Table 2.

Association of adversity during sensitive periods with inflammatory biomarkers.

Timing of adversity exposure	M1/M2-like phenotype			Endotoxin tolerance phenotype			Pro-inflammatory cytokines			Anti-inflammatory cytokines
	β	t-test (df)	p	β	t-test (df)	p	β	t-test (df)	p	β	t-test (df)	p
Early life
Unadjusted	0.337	2.265 (40)	0.029*	0.066	0.420 (40)	0.676	0.126	0.909 (51)	0.368	0.045	0.321 (51)	0.750
Adjusted model 1	0.308	2.094 (36)	0.043*	0.064	0.384 (36)	0.703	0.137	1.121 (47)	0.268	0.040	0.311 (47)	0.757
Adjusted model 2	0.311	2.114 (37)	0.042*	0.084	0.511 (37)	0.612	0.184	1.486 (48)	0.144	0.062	0.442 (48)	0.660
Pregnancy
Unadjusted	0.332	2.226 (40)	0.032*	0.150	0.962 (40)	0.342	−0.185	−1.344 (51)	0.185	−0.333	−2.519 (51)	0.015*
Adjusted model 1	0.404	2.764 (36)	0.009**	0.180	1.061 (36)	0.296	−0.069	−0.529 (47)	0.599	−0.230	−1.745 (47)	0.087†
Adjusted model 2	0.324	2.049 (37)	0.048*	0.203	1.187 (37)	0.243	−0.123	−0.933 (48)	0.355	−0.294	−2.081 (48)	0.043*
Total lifespan
Unadjusted	0.162	1.012 (38)	0.318	−0.341	−2.239 (38)	0.031*	0.188	1.323 (48)	0.192	0.045	0.311 (48)	0.757
Adjusted Model 1	0.054	0.315 (34)	0.755	−0.410	−2.417 (34)	0.021*	0.135	1.086 (44)	0.283	0.001	−0.001 (44)	0.999
Adjusted Model 2	0.051	0.301 (35)	0.765	−0.415	−2.544 (35)	0.016*	0.314	2.398 (45)	0.021*	0.101	0.661 (45)	0.512

**p ≤ 0.01, *p ≤ 0.05, †p ≤ 0.10. Separate linear regression models were fitted, each specifying one inflammatory marker as the outcome and the specified life adversity factor as the independent variable. The anti-inflammatory and pro-inflammatory aggregates include the average of the blom-standardized scores for IL-1RA, IL-4, IL-10, and IL-1B, IL-6, TNF-a, and CRP respectively. Phenotype definitions are given in the methods. Adjusted model 1 included: age, BMI, and Hispanic ethnicity and current antidepressant use as covariates; adjusted model 2 included: high school education, family poverty, and being US born. Three individuals were missing data for total life adversity.